SuperDial - FDE (Copy)

Deploy Talent • San Francisco, CA

30d • $120,000 - $185,000

About The Position

We’re looking for an Applied AI Engineer to help us unlock the full potential of
large language models in real-world healthcare workflows. If you love
experimenting with LLMs, working closely with product and engineering teams, and
obsessing over real user outcomes—this role is for you.
About the Role:
Design, iterate, and optimize prompts that drive LLM behavior in production
settings—spanning classification, summarization, structured extraction, and
workflow automation.
Collaborate with engineers, product managers, and customers to translate
domain problems (e.g. prior auth, claim appeals, denials) into robust
LLM-powered solutions.
Build and maintain internal tooling for prompt versioning, testing, and
performance evaluation (accuracy, latency, cost).
Fine-tune and evaluate models where needed using structured feedback and
real-world data.
Analyze edge cases and failure modes; design prompt strategies to improve
resilience and alignment with business logic.
Stay current on the LLM ecosystem and incorporate best practices into our
architecture.

Requirements

• 2+ year of experience in a software engineering and LLM-centric role.
• Deep understanding of prompt design patterns and LLM capabilities/limitations across GPT-4, Claude, Mistral, etc.
• Hands-on experience with LangChain, LlamaIndex, PromptLayer, or similar frameworks.
• Strong Python skills and comfort working in a codebase (bonus: experience with backend workflows or vector DBs like Weaviate/Pinecone).
• Obsession with quality—you’re comfortable writing tests, measuring performance, and debugging model behavior with structured rigor.

Nice To Haves

• Exposure to healthcare data, terminology (EHRs, RCM, FHIR), or compliance-aware deployments.

Responsibilities

• Design, iterate, and optimize prompts that drive LLM behavior in production settings—spanning classification, summarization, structured extraction, and workflow automation.
• Collaborate with engineers, product managers, and customers to translate domain problems (e.g. prior auth, claim appeals, denials) into robust LLM-powered solutions.
• Build and maintain internal tooling for prompt versioning, testing, and performance evaluation (accuracy, latency, cost).
• Fine-tune and evaluate models where needed using structured feedback and real-world data.
• Analyze edge cases and failure modes; design prompt strategies to improve resilience and alignment with business logic.
• Stay current on the LLM ecosystem and incorporate best practices into our architecture.

Benefits

• Build AI that has real impact - help reduce burnout and drive efficiency in healthcare systems that affect millions.
• Work with a small, high-velocity team obsessed with shipping.
• Shape the future of our AI stack from the ground up.
• Competitive compensation + meaningful equity + healthcare benefits.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level
• Education Level: No Education Listed